PrepShorts · Study sheet · Class 7 Mathematics · Chapter 3, Finding Common Ground
Chapter 3 · Finding Common Ground
Conjecture and generalisation: what mathematicians mean by those words
This video could not be loaded. Reload the page to try again.
Sign in with Google10 min.
Keep your place in this chapter — sign in, it’s free.Sign in
This one is not about factors. It is about the status a claim carries, and about saying which status you are claiming.
The idea
This chapter quietly teaches something that is not about factors at all: that a mathematical claim carries a status, and that saying which status you are claiming is part of making the claim. A conjecture is something nobody has yet argued for, and a single awkward example ends it. A general statement is one that has been argued to cover every case, which is why a letter appears the moment the chapter wants one. And the chapter practises what it preaches on its own flagship result: it works one pair on the page, hands the reader three more to explore, names the pattern across all four — the HCF times the LCM keeps landing on the product — and then still refuses to promote the finding. It says the observations point that way, prints a hint, and asks the reader to supply the argument. Then it does something braver still, and asks whether the same claim survives being stretched to three numbers, without saying.
What you should be able to do
- Say what makes a claim a conjecture rather than an established fact
- Refute a stated conjecture by producing one instance where it fails
- Say what a general statement claims, and what would be needed to earn it
- Explain why the LCM of two numbers can never exceed their product
- Show, on a worked pair, that the product divided by the LCM is the HCF
- State how a pair's HCF, its LCM and its product are related, and state it with the same caution the chapter uses
- Follow the chapter's hint far enough to explain why the relation holds for two numbers
- Test whether the relation survives three numbers, and report honestly what the test shows
Words to know
| Term | Definition in one line | First introduced |
|---|---|---|
| conjecture | a claim put forward before anyone has argued for it or checked it | printed in single quotes in a side box, §3.1, Part II, p.52, and again in the SUMMARY, p.65 |
| counterexample | one instance in which the claim fails, which is enough to end it | printed in the same side box, §3.1, Part II, p.52 |
| verification | checking a claim against cases | printed in the same side box, §3.1, Part II, p.52 |
| general statement | a claim asserted to hold in every case of the kind described | printed in bold in §3.3, Part II, p.58 |
| generalisation | the act of arriving at such a claim | printed in bold in §3.3, Part II, p.58, and in the SUMMARY, p.65 |
| proof | an argument that settles a claim for every case at once | printed in §3.1, Part II, p.52 and again in §3.3, Part II, p.63; the verb prove is printed in §3.3, Part II, p.62 |
| explanation | the chapter's gentler word, offered alongside proof | printed in §3.3, Part II, pp.59 and 63 |
| property | a relation that holds between quantities attached to the same numbers | printed in the subheading "Property Involving both the HCF and the LCM", §3.3, Part II, p.62 |
| relation | the link the chapter says ties a pair's product to its HCF and its LCM | printed in §3.3, Part II, p.62 |
| theorem | a claim that has been proved | an added term; this chapter never prints it, stopping at proof and general statement (all twenty pages, pp.47–66, read) |
Where people slip up
- "A conjecture is a wrong guess." It is an unsettled claim. Some conjectures are true and unproved; Anshu's happens to be false. Nothing about the word says which. The chapter's second box is explicit that the reader should expect to make conjectures of their own, which would be strange advice if the word meant error.
- "Lots of examples make a claim true." They make it worth trying to prove. The chapter works one pair through the HCF-times-LCM relation, sets three more for the reader, names the pattern it sees across all four, and still declines to state it as settled.
- "One counterexample only weakens a claim." It ends it, as a general claim. The claim may survive in a narrower form — Anshu's holds only for powers of a single prime, and fails even for numbers built from the same primes — but the sentence as stated is finished.
- "HCF × LCM = product holds for any number of numbers." It does not. This is the single most likely error an explanation will make here, because the two-number version is so tidy and because the chapter's own question invites the leap. 2, 3 and 4 settle it. Present the three-number case as an experiment with a surprising outcome, and never as a stated rule.
- "The LCM could be bigger than the product if the numbers are big enough." It could not, ever: the product is already a common multiple, and the LCM is the least of them. Size is irrelevant, which is why the chapter does not let the reader stop at examples but goes on to ask for a reason or a proof.
- "The chapter proves the relation." It gives a hint towards a proof and asks the reader for one. If the explanation supplies the argument — and it should — it must say that this is the explanation doing what the book asked, not the book speaking.
Ask your teacher a person
Your teacher reads this and writes back, usually within a day. For an instant answer, use Ask the video in the sidebar.
Your class sees the question and the answer. Only your teacher sees that it was you.
No questions on this topic yet.
Worked answers to this chapter’s exercises
Transcript1,385 words
Here is a claim about numbers. Bigger numbers have longer prime factorisations. Which sounds reasonable. Eight is two, two, two — three primes. A hundred is two, two, five, five — four primes. Bigger, and longer. And it keeps working. Check every pair below two hundred and it holds on thirteen thousand of them. So is it true? Not yet. Right now it is a claim nobody has argued for, and there is a word for exactly that.
A conjecture. A conjecture is not a wrong guess. It is an unsettled one. Some turn out true, some turn out false, and the word says nothing about which. What it does say is that the work is not finished. So let us try to finish it. There are two ways that can go. You can argue that it has to hold every time. Or you can find one case where it does not.
Ninety-six. Two, two, two, two, two, three. Six primes. A hundred and twenty-one. Eleven times eleven. Two primes. A hundred and twenty-one is the bigger number, and it has the shorter list. That is a counterexample, and one is enough. Not weakened, not mostly true. Finished. Thirteen thousand agreeing cases did not save it. Because a claim about every pair is only ever as strong as its worst pair. Now the natural next move. Patch it.
Maybe the trouble was comparing an eleven against twos and threes. So compare numbers built from the same primes. Forty is two, two, two, five. Fifty is two, five, five. The same two primes in both. Fifty is the bigger one. And fifty has the shorter list. The patch fails as well, and it fails at forty and fifty, which is not a hard place to look. There is a version that survives, but it is much weaker. If one number divides the other, the multiple has at least as many primes.
And that one you can see the reason for. The multiple is the divisor times something, and that something brings its own primes with it. Which is the whole difference. Not more examples. A reason. So there is another kind of claim, at the other end. Here is one. Whenever one number divides the other, their highest common factor is the smaller of the two. Six and eighteen give six. Five and twenty give five. Three and twelve give three.
But this is not being offered as something to go and check. It is offered as something that already covers every case, and that is called a general statement. The difference between the two is not confidence. It is what is owed. A conjecture owes you either a reason or a counterexample. A general statement claims the reason has already been given. And saying which of the two you are making is part of making it.
One move does more than anything else to turn the first kind into the second. Use a letter. Call the smaller number n. If the smaller divides the larger, then the larger is a multiple of n. So take n and five n. What is their highest common factor? It is n. n divides n. n divides five n. And nothing bigger than n divides n. That argument did not check a single pair, and it covers all of them at once.
Which is what a letter is for. It is not shorthand for a number you happen not to know. It is a way of saying every case, in one line. Now a harder claim, handled properly this time. Take two numbers. Which is bigger — their lowest common multiple, or the two of them multiplied together? Six and eight. Multiplied, forty-eight. Lowest common multiple, twenty-four. The product wins. Four and nine. Multiplied, thirty-six. Lowest common multiple, thirty-six. A draw.
Try as many as you like and the multiple never wins. But that is examples again, and we know what examples are worth. So here is the reason, and it takes one sentence. The two numbers multiplied together is itself a common multiple. And the lowest common multiple is the lowest of all of them. So it cannot beat one of them. Settled. And the draw happens exactly when the two numbers share nothing.
That comparison leaves something behind, though. When the product wins, by how much does it win? A hundred and five, and ninety-five. A hundred and five is three, five, seven. Ninety-five is five, nineteen. Their lowest common multiple takes every prime at its larger count. Three, five, seven, nineteen. One thousand nine hundred and ninety-five. Their product is nine thousand nine hundred and seventy-five. So divide one by the other. Nine thousand nine hundred and seventy-five, divided by one thousand nine hundred and ninety-five, is five.
The product is the lowest common multiple, times five. So where did the five come from? Three more pairs, with nothing given away in advance. Forty-five and one hundred and five. Lowest common multiple, three hundred and fifteen. The product divided by it leaves fifteen. Two hundred and seventy-five, and three hundred and fifty-two. Lowest common multiple, eight thousand eight hundred. Leftover, eleven. Two hundred and twenty-two, and three hundred and seventy. Lowest common multiple, one thousand one hundred and ten. Leftover, seventy-four.
Fifteen. Eleven. Seventy-four. And the five from before. Now put those four numbers next to the pairs that produced them. Fifteen is the highest common factor of forty-five and one hundred and five. Eleven is the highest common factor of the second pair, seventy-four of the third, and five of the first. Every single time, the leftover is the highest common factor. So here is what those four pairs point towards.
The highest common factor, times the lowest common multiple, equals the two numbers multiplied together. And notice what has just happened. We have four agreeing examples and a very tidy pattern. By the standard set at the start of this video, that is a conjecture. A good one. Worth stating, and worth trying to prove. But this whole video so far has been about why agreeing examples are not a proof.
So we do not get to stop here. We owe a reason. And the reason is short, once you look at one prime at a time. Pick any prime. Say the first number holds three copies of it, and the second holds five. The highest common factor takes the smaller count. Three. The lowest common multiple takes the larger count. Five. And the two numbers multiplied together holds both lots together. Three plus five. Eight.
Now put the first two side by side. Three and five. Added, that is eight. The smaller count plus the larger count is just the two counts added. That is the entire argument. Every prime balances the same way, so the two sides match exactly, for every pair there is. It is not a pattern any more. Which raises the obvious question. What about three numbers? Take two, three and four.
Their highest common factor is one. Their lowest common multiple is twelve. One times twelve is twelve. And two times three times four is twenty-four. So no. Twice too small, and the claim does not survive. But it is not simply false either. Try five, seven and nine. Highest common factor one, lowest common multiple three hundred and fifteen, and the product is three hundred and fifteen. That works. And checking every triple up to thirty, it works exactly when no two of the three share a factor.
Which is stricter than it sounds. Six, ten and fifteen have nothing common to all three, and it still fails, because six and ten share a two. So what was all of this really about? Every claim you meet carries a status, and the status is part of the claim. Some are conjectures. Put forward, not argued, still waiting. Some are finished, by a single example that goes the wrong way.
And some are general statements, which means somebody has given a reason that covers every case at once. Making conjectures is not the mistake. It is the ordinary first move, and you will make them constantly. The mistake is losing track of which one you are holding. Because a pattern is a question. A reason is the answer.
Where this fits
Taken from the notes each video was made from, not from the reading order — these are the ideas this one rests on and the ones that later rest on it.
Builds on
- HCF: take the fewest occurrences of each primeClass 7 · Ch 3, Finding Common Ground
- LCM: take the most occurrences of each primeClass 7 · Ch 3, Finding Common Ground
- What HCF and LCM do for consecutive, even, and co-prime numbersClass 7 · Ch 3, Finding Common Ground
Comes up again in
- Divisions that never endClass 7 · Ch 4, Another Peek Beyond the Point
Either side of this one
- Getting the HCF and the LCM out of one division ladderClass 7 · Ch 3, Finding Common Ground
- Multiply as whole numbers, then count the decimal digitsClass 7 · Ch 4, Another Peek Beyond the Point